On Tue, 27.09.11 18:21, Stefan G. Weichinger ([email protected]) wrote: > > Am 27.09.2011 16:19, schrieb Lennart Poettering: > > > Also see this blog story I posted yesterday with a longer > > explanation: > > > > http://0pointer.de/blog/projects/inetd.html > > > > (Sorry if that loads very slowly only. The server is currently under > > pressure due to the PA 1.0 release) > > Thanks for that blog-entry. I read through it and tried to apply that to > my amanda-related systemd-files. Unfortunately I didn't make any > progress, I still get amanda failing after its so-called estimate-phase. > > I assume that things aren't so easy when I read the amanda-docs about > how it uses ports ... > > Using it via the intermediate layer of starting xinetd.service is > working fine so far, so I think I leave this to more experienced people ;-)
There's some borked software around which when it finds getppid() == 1 follows a different code path on init then otherwise, in the ill-fated attempt to detect whether a daemon is started from the old inittab instead of on an init script. You might want to check whether your software is doing something like tht. Lennart -- Lennart Poettering - Red Hat, Inc. _______________________________________________ systemd-devel mailing list [email protected] http://lists.freedesktop.org/mailman/listinfo/systemd-devel
